Search the Site

Episode Transcript

JENA: Julie, you know, I’ve got two loving kids. You’ve got kids too, right?

KANFER: Yes, I’ve got two. I’m not sure how loving they always are, (BJ^laughs) but I’ve got two also.

JENA: You know, this is being recorded, right?

KANFER: Oh yeah. It’s fine.

JENA: I want to tell you about a study that I did a while ago. This was probably one of my first ideas I was in grad school, I was living in Chicago, and I was interested in this question of the impact of children on marriages. And at that time, I thought, is there a way to find a natural experiment whereby some couples by chance are exposed to more kids and other couples are not?

It turns out, there was a way. The challenge with that kind of study is that you can’t simply compare couples with more kids to couples with fewer kids, because those couples are different for all kinds of reasons. What I needed was a situation where some couples are exposed to more kids randomly — maybe because they gave birth to twins. I’m Bapu Jena, and this is Freakonomics, M.D. Today on the show, do twins cause their parents to get divorced? And how might other unexpected events impact whether a couple stays together or not?

JENA: Now you might be thinking, Well, in 2023, many couples who have twins have twins because of I.V.F.

KANFER: Right. So is that still considered random?

You may be wondering the same thing as my producer Julie Kanfer and with good reason. Using assisted reproductive technologies like I.V.F., or in-vitro fertilization, does raise the odds of becoming pregnant with twins, or any multiples. And these technologies are more popular than ever: according to C.D.C. data, their use has more than doubled over the last decade. But people who go through fertility treatments are not necessarily the same as people who don’t, or can’t. So, to get to the bottom of how children really impact their parents’ marriage, I had to search for a time before assisted reproductive technology was so commonly used.

KANFER: So you had to go all the way back in time to 1980? 

JENA: Yeah, I don’t even want to know what year —

KANFER: Who can even remember?

JENA: You, you probably weren’t — I was barely alive in, in —

KANFER: I was, uh, a year away from being born.

JENA: So what you have to do is you have to look way back in time to a world when I.V.F. wasn’t really around. And you look at couples who have twins spontaneously, so not through I.V.F. And in that scenario, you could look at couples who have twins and compare them to couples who don’t have twins, and essentially ask two questions. First of all, what is the impact of having twins on the total number of children that a couple has? And then the second thing you can do is look at what is the effect of twins on divorce by looking at divorce rates among couples who have twins versus couples who don’t. And I started looking at census data from the 1980 U.S. census. I had data on about 800 to 900,000 women who had at least one birth and then about 14,000 women who had twin births. And I looked at married couples and whether or not the rate of divorce differed based on the presence of twins in that household.

KANFER: What did you find?

JENA: I found a couple of things. So first is that couples who had twins were more likely to be divorced by the time of the census survey. Among women who had been married the rate of divorce at the time of the survey was about 13.7 percent among women who had twins and 12.7 percent in women who did not have twins. So that’s roughly a 10 percent increase in the rate of divorce.

KANFER: Were you surprised by that finding?

JENA: I was a little bit surprised in the following sense. If you look at the raw empirical relationship between couples who have kids versus couples who don’t, couples who have kids are less likely to get divorced than couples who don’t have kids. And, again, there’s a selection bias that really prevents us from saying anything causal about whether or not kids are protective of marriage. So in that sense, it was contradictory to what sociologists would’ve studied at that point. but it’s not surprising if you think about what twins entail. It’s not just the presence of an additional child, but it’s the presence of an additional child at the same time as another child. And, that period of time where you have two kids who are born versus separated by, let’s say two or three years probably matters a great deal for stress to the relationship. But that brings me to the second point. The stress of having twins in the way I just described, you would think then would lead to divorce early on, shortly after couples have twins. But that’s not actually what I found. What I found is that most of the divorces are occurring sometime later on after kids are 8 or 9-years-old.

KANFER: So what’s going on there? Is it just that maybe when they’re super little you kind of need the help? You need each other more to participate in the intense childcare that toddlers and young children require.

JENA: Yeah. I think that’s probably it. I mean, it takes time for couples to make these sorts of decisions. And the cumulative effect of having perhaps one more child than you had planned, that grows over time. So there’s two other things that I’ve found, and maybe it’ll help us understand what exactly is driving these findings. I looked at whether or not the sex of the twins mattered. And what you find is that couples who had at least one twin girl had higher rates of divorce than couples who had twin boys. And in fact couples who had twin girls had the highest rates of divorce compared to couples who had twin boys. And couples who had twin boys are pretty similar to couples who did not have twins. So there’s something specific about having a twin girl or two twin girls that is particularly destabilizing to the marriage. And the second finding was that the effect of twins on divorce was larger in, women who were less educated and had lower income. That maybe speaks to the stress associated with having to care for two kids or an additional child that you hadn’t planned on, that comes with having lower education and lower income, because children are expensive and they take time to raise.

KANFER: So there was also a 1988 study published in the American Journal of Sociology, just on the point of boys versus girls as part of the twin set. This study found that couples with sons were 9 percent more likely to stay together than couples with daughters. And the odds of staying married increased by 9 percent with each additional son that was born. And one of the theories is that for fathers, the obligations and attachments to children were greater if they had sons.

JENA: And so Julie, there have been a number of other papers that have looked at that type of question, and maybe that partly explains why we see that differential rate of divorce in couples who have twin girls versus couples who have twin boys. There is a famous economic paper called “The Demand for Sons” and it showed that shotgun marriages were more like, to occur if a couple had, a boy versus a girl. We do know that there is a preference for sons in other countries, and that’s been widely shown. It’s less clear whether we would expect to see that in the U.S. given it’s more educated and then higher income. But you still see some evidence of that kind of playing out. And this study is a good example, but there’s other ones like it.

Having twins unexpectedly falls into a larger category of reasons marriages can fail. After the break: why do unexpected events — good or bad — increase the risk of divorce? I’m Bapu Jena and this is Freakonomics, M.D.

*      *      *

There’s a lot of research across many fields that tries to answer questions about divorce. To me, one of the most influential works comes from economics, and was published by the University of Chicago economist and Nobel laureate Gary Becker in the Journal of Political Economy back in 1977. It’s called, “An Economic Analysis of Marital Instability,” and it explores a simple but ultimately very challenging question: Why do people get divorced? Back to my chat with Julie.

JENA: The underlying idea in that paper was that divorce occurs because there are deviations from expectations that people have about what their lives will look like together. And that I think has a fundamental insight because for the twin study, twins are random, right? Couples don’t expect to have twins and there are lots of things that can cause deviations from what you expect in your life. It could be losing a job, which is a negative shock. We know that that is associated with high rates of divorce. It could be a negative health event like cancer. Again, we know that there’s higher rates of divorce among people who have cancer. It could be positive shocks, could be something like getting a promotion. Those set your life in a different trajectory than you had otherwise planned. And again, those deviations from expectations, positive or negative, are what Gary Becker, and his colleagues argued are at least one driver of divorce. And that does actually have relevance to health because there are negative health effects that we know are associated with divorce but you could imagine other things that are sort of positive effects on your health that could also lead to divorce.

KANFER: Yeah. So what are some of those?

JENA: All right, so I, I’ve got two ideas and neither one of them are very well fleshed out, but imagine that you have a new medical technology that comes out and that medical technology allows you to lose weight. Let’s say, it’s like some form of weight reduction surgery. The arrival of that technology would change the dynamic of marriage. Again, two people decide to get married based on an expectation of a lot of factors that they think will gel over the course of their lifetimes. If someone loses a dramatic amount of weight because of a new surgical technique, that sort of deviation, in this case would be a health shock, could destabilize marriage. Same thing might be true for, new cosmetic surgeries that sort of change people’s perceptions of themselves and how they look. They change the dynamic of the marriage. So new medical technologies would have this sort of effect.

KANFER: Well, in 2022, the Annals of Surgery Open, there was a study published that found adults who are not married and get weight loss surgery are more than twice as likely to get married within five years, which is interesting. So people are getting this surgery and then, compared to the general population, they become more likely to get married after having the surgery. But adults who are already married when they get weight loss surgery are more than twice as likely to get divorced.

JENA: There’s two findings it seems like. So one is that adults who are not married, if they get weight loss surgery are more likely to get married within five years. (JK^Yes. ) And the question is, did the weight loss surgery cause that to occur? Or are these individuals who are looking to get married and they think that losing weight will help them in that prospect, and so they decide to get weight loss surgery. Sort of different mechanism that’s at play there. Similar thing could be true for the divorce channel. If you look at adults who are married and one spouse gets bariatric surgery, and that couple then ends up being more likely to get divorced, did they get bariatric surgery knowing the marriage was not going well and that they would probably want to go back out on the dating market at some point? Hard to know what drives what. That’s why I think looking at instances where the surgery or medical technology is coming in really randomly is as important, right?

KANFER: Yeah. How could you do that?

JENA: You could look at the geographic diffusion of bariatric surgery. It’s probably the case that some medical centers started adopting the surgery earlier rather than later. And you could focus on individuals who were overweight or who are at high predicted risk of seeking bariatric surgery and look before and after in those markets what happens to rates of divorce in those couples, and presumably the effects that you’d see would be driven only by those couples in which one of the people decided to get bariatric surgery. But the key thing is you’re looking for situations where people are exposed somewhat randomly to the option of getting bariatric surgery versus not.

KANFER: Well, we also discovered that there was one medical event that could have maybe a stabilizing effect. This paper published in N.B.E.R., National Bureau of Economic Research in 2017 found when Medicaid was expanded there were fewer “medical divorces”, which is when a couple splits up because of medical bills, basically. But when Medicaid was expanded, it reduced the incidences of those kinds of divorces.

JENA: I remember seeing that paper. It’s an interesting one. I think the basic idea is that if you have a couple who’s married and one person has an adverse medical event that’s gonna be costly and it’s gonna rack up medical bills, they don’t want that to deplete the assets of the other person to whom they’re married, and that would be called a medical divorce. Now, I wouldn’t have expected to see an impact of Medicaid expansions to reduce medical divorce. And the idea is that the expansion of Medicaid for these groups of people who have lower income reduces that financial instability.

KANFER: We’ve talked a lot about some other possible triggers that would cause divorce, but what about things like this Medicaid expansion that seem to have the opposite effect?

JENA: So interventions or policies that reduce that destabilizing effect could in theory reduce the rates of divorce. And it could be a healthcare insurance expansion. It could be disability insurance; it could be better unemployment insurance. One prediction here would be that if you look at couples who have access to better unemployment insurance that we should observe lower rates of divorce there because the destabilizing event of losing a job can be countered by the stabilizing event of getting unemployment insurance. And then again, you know, I’m a health economist, this is what I do but I think there’s a lot of ways in which health and deviations from what we expect our health to look like can impact our lives and an important part of many people’s lives is their marriage.


Like divorce itself, this topic is tricky to navigate from a research perspective. I wanted to talk about my early research on twins because it uncovered interesting findings not just about divorce, but about the ways that unexpected, random things can affect our lives and our health.

That’s it for today’s show — and for the show itself, for a little while. I’ve got some news to share: Freakonomics, M.D. is going on hiatus while I get ready for my first book to be published this July. It’s called Random Acts of Medicine, and it’s co-authored with Dr. Chris Worsham, one of my longtime research partners. The book is very much in the spirit of this show, drawing on lessons from economics to help you better understand how doctors think, how medicine works, and how it could all be done better. I’m really proud of Random Acts of Medicine, and if you like this show, I think you’ll love the book.

Now, I have a small favor to ask: If you know you plan to buy the book, could you think about pre-ordering it? Pre-orders help booksellers and publishers gauge interest in a book and your pre-order could make a big difference in helping others find their way to it when it comes out in July. So that’s my request. Stay tuned and you’ll hear more from me soon. Thanks to my producer Julie Kanfer for joining me today, and thanks to you, of course, for listening and supporting Freakonomics, M.D.

*      *      *

Freakonomics, M.D. is part of the Freakonomics Radio Network, which also includes Freakonomics Radio, No Stupid Questions, and People I (Mostly) Admire. All our shows are produced by Stitcher and Renbud Radio. You can find us on Twitter at @drbapupod. This episode was produced by Julie Kanfer and mixed by Eleanor Osborne. Lyric Bowditch is our production associate. Our executive team is Neal Carruth, Gabriel Roth, and Stephen Dubner. Original music composed by Luis Guerra. If you like this show or any other show in the Freakonomics Radio Network please recommend it to your family and friends. That’s the best way to support the podcasts you love. As always, thanks for listening.

JENA: My hope for you, Julie, is that you live a long, happy life and that any medical technology that you need is available to you.

KANFER: Thank you so much, Bapu. Same to you.

JENA: Yeah. Thank you. Thank you.

Read full Transcript



Episode Video